Multi-site Site Acceptance Criteria: What Changes at Scale
This composite field diary follows a multi-site smart-lock handover across a border. The phrase site acceptance criteria multi site may sound like a document label; on this assignment, it was a working problem. A door that passed inspection at one location created confidence, but it did not create a transferable acceptance record. By the third site, the programme team had learned to separate the common test, the local requirement, and the evidence that allowed each decision to stand.
Direct answer. Short version — site acceptance criteria multi site is treated here as site acceptance criteria multi site first. site-acceptance-criteria comes second. The evidence for site acceptance criteria multi site is named in the sections below. A site acceptance criteria multi site sets out the maintenance record for a system in service. site acceptance criteria multi site carries the early-warning signal for site acceptance criteria multi site.
- Set the interval — define how often site acceptance criteria multi site is inspected
- Set the trigger — name the condition that forces a change to site-acceptance-criteria
- Track the number — keep one metric that shows departure board is healthy
- In site acceptance criteria multi site, this sequence is what site acceptance criteria is checked against.
The diary contains no named client or supplier claim. It records the operating pattern that emerged while an illustrative portfolio moved from installation into post-award operations. The facility manager needed more than a signed inspection sheet. They needed to know which rules applied, who could approve an exception, what evidence remained valid after the team crossed a border, and what happened when a later site differed from the first.
The practical change was subtle but important. At one site, site acceptance criteria were a set of checks performed on a door. At portfolio scale, they became a controlled handover system. The physical test remained short; the surrounding governance became more explicit.
Departure board: the acceptance rule had to travel
Before departure, the programme team wrote one sentence at the top of the acceptance pack: no door enters normal service until its agreed handover checks are complete and the evidence is attached to the correct object record. That sentence replaced the vaguer instruction to “complete commissioning.” It also established the distinction between a technical result and an administrative result. A working lock with an incomplete record is not an accepted door. An incomplete record beside a working lock is an open exception, not permission to imply that the check occurred. Every commercial claim here has to be checked against the smart door locks.
The programme treated the acceptance rule as a travel document. It identified the object schedule, the applicable test sequence, the evidence required, the authority for exceptions, and the version in force at each location. This mattered when the local team used different terminology for the same door type or when a handover document was translated. The meaning travelled through the matrix and record identifiers; the wording in the local annex could change.
The physical survey still came first. For a new installation, the team used the Office Smart Lock Site Survey as a reference for door condition and access context. Where an existing door was being adapted, the Smart Lock Retrofit Site Survey provided a separate route. Conflating a new fit with a retrofit would have made the acceptance matrix look uniform while concealing different construction conditions.
One further rule was agreed at the departure desk: a site could not borrow evidence from another site merely because both used a similar lock. The handover record had to identify the door or unit, the model and variant, the check performed, and the observation that supported the result. Portfolio consistency would come from using the same decision structure, not from asserting that every location behaved identically.
Day one: a clean door with an incomplete file
The first site looked promising. The door closed cleanly, the handles operated as expected, and the installed unit had a legible label. The folder was less complete. It contained a photograph of the lock and an installation note, but it did not show who had tested the access path or what observation had been made. It also mixed the door identifier from the schedule with a room label, making the handover record ambiguous.
The team did not convert the photograph into proof. The technician recorded the object identifier, the model and variant from the label, the test time, the symptom or condition in plain words, and the measured observation. Where the check had a diagnostic sequence, the record named each check and what it ruled out. That last column is what separated a sequence of actions from usable evidence.
| What the handover showed | What the record still lacked | Decision at the door |
|---|---|---|
| Door closed and handles moved | Dated test record linked to the scheduled object | Hold administrative release |
| Lock body visible and labelled | Model and variant recorded against the correct row | Correct the object record |
| Credential was said to work | Credential state before and after the test | Repeat the defined access check |
| No apparent physical defect | Local annex and exception route named | Attach the approved annex |
This was not a remote-monitoring test or a performance claim. It was a handover decision at the object. The field team followed the physical state, door alignment, latch engagement, configured access route, and administrative state in the approved order. If a lock did not move, the team followed the written fault sequence rather than jumping to a replacement. The rationale is documented in Site Acceptance Criteria: A Field Worksheet to Complete, which can be used for the door-level capture.
The day's lesson was that a visible success and a defensible acceptance are different things. Visibility helps, but it does not tell the next owner what was checked, what the check meant, or which door it covered.
Day two: the local annex changed the evidence
At the next site, the common core remained recognizable, but the evidence sources did not. The local coordinator used a different approval route, a different document language, and a different system for retaining technician images. The team did not force these differences into a generic tick box. It recorded them in a local annex attached to the same portfolio matrix.
The annex had to name the source and the owner. It also had to state which part of the common test the local evidence completed and which part still required observation at the door. A local approval document was not allowed to stand in for a mechanical or access test unless the approved matrix explicitly allowed that substitution.
Local building, access, emergency, employment, and data-handling questions also cannot be settled by a generic overseas template. The responsible buyer must identify the applicable authority and retain the relevant decision.
CONFIRM-WITH-LOCAL-BUILDING-AND-ACCESS-AUTHORITY
Fire-egress, accessibility, and emergency arrangements need the same discipline. The handover packet must not carry an unsupported “approved” label. It should identify who reviewed the installed arrangement, what evidence they issued, and which part of the door handover that evidence covers.
CONFIRM-WITH-LOCAL-FIRE-SAFETY-AND-ACCESSIBILITY-AUTHORITY
The risk owner may also request an insurer-specific view of the evidence. This diary makes no assumption about whether an insurer will accept a record, what format it requires, or who must receive it.
CONFIRM-WITH-LOCAL-INSURANCE-AND-RISK-OWNER
A dated product page was added as the reference for any product-specific assumption in the test sheet. The team did not infer a fallback method, credential type, power input, or administrative capability from a neighbouring model. Confirm on the dated product page for the specific SKU before using any of those assumptions in a pass condition.
Portfolio acceptance matrix: three layers of authority
By the second evening, the acceptance pack had been restructured around a portfolio acceptance matrix. The team stopped describing one list of tests and began separating the common core, the local annex, and the exception schedule. This made ownership visible. It also kept a local variation from rewriting the purpose of the handover.
| Layer | What it decides | Portfolio owner | Local owner | Required evidence |
|---|---|---|---|---|
| Common core | The physical, access, administrative, and record checks applied to every door in scope | Portfolio operations lead | Site acceptance lead | Door-level result linked to the object record |
| Local annex | The evidence required to interpret the core in that location | Programme governance owner | Local compliance or operations owner | Named authority, source document, scope, and review record |
| Exception schedule | The treatment of a failed or incomplete check | Contract and delivery owner | Named action owner | Reason, owner, required closure evidence, and status |
| Change record | Any approved change to the matrix or a site-specific interpretation | Change controller | Local confirmer | Version, reason, affected doors, and approval |
The hierarchy solved a recurring argument. The common core prevented one site from quietly omitting a basic check. The local annex prevented the core from pretending that one jurisdiction’s evidence always looked the same. The exception schedule prevented a failed check from disappearing into an email thread.
flowchart TD
A[Receive versioned handover pack] --> B{Common core complete}
B -->|No| C[Hold door and name missing evidence]
B -->|Yes| D[Check local annex]
D --> E{Local annex accepted}
E -->|No| F[Open local exception]
E -->|Yes| G[Make acceptance decision]
F --> G
G --> H{All release conditions closed}
H -->|No| I[Set owner and closure evidence]
H -->|Yes| J[Release site acceptance record]
The flow deliberately placed release after evidence closure. A conditional item could proceed only under the rules in the matrix. It could not be converted into acceptance merely because the programme had a handover date.
At the door: running the common-core sequence
The common-core sequence gave the field team a stable order of work. It started with the door in its approved operating state. The technician checked the door, frame, latch, and handle condition before treating the installed unit as the source of an issue. This separated a door or latch condition from a lock condition and prevented a unit swap from becoming the first diagnostic step.
Next, the technician tested the access path that the approved configuration actually assigned to that door. The record captured the credential or identifier tested, its state before the test, and its state afterwards. A credential working at another door did not prove that the intended door was correctly provisioned. The record also noted what the check ruled out: physical obstruction, wrong object record, stale configuration, or an access-path mismatch.
The administrative check then examined the state visible to the local owner. Its depth followed the scope of the handover and the capabilities assigned to the selected SKU.
The technician then completed the approved fallback and resilience checks. If the selected variant includes a fallback route, its operation and record were tested according to the matrix. Where a check was not applicable, the reason was recorded rather than left blank. “Not applicable” without an authority can conceal an untested requirement; “out of scope under the approved matrix” creates a traceable decision.
Finally, the technician recorded the result, the time, the checks run, and what each check ruled out. If a part had been fitted or a unit exchanged during correction, the old unit was retained under the programme’s evidence rule. The link to the complete evidence pattern appears in The Evidence a Site Acceptance Criteria Record Needs. The point was not to create a long report. It was to preserve a sequence another person could follow.
Between site one and site twenty: what scale changed
Scale changed the source of risk. At one site, the main danger was a missed check on one door. At twenty sites, the greater danger was a familiar check being treated as universally understood. A local team might use the right words, record the same photograph, and still apply a different pass condition unless the matrix made the condition explicit.
The programme therefore separated test definition from test execution. The portfolio team owned the question being asked, the evidence class required, and the decision rule. The local team owned the observation, the local annex, and the people or records supporting the result. This arrangement did not centralise every technical task. It prevented local variation from silently changing the portfolio meaning.
The second change was versioning. Each site received a stated matrix version and a list of affected door identifiers. A later interpretation was not applied retrospectively without a change record. If an approved clarification affected a completed door, the change controller could identify whether that door required review. This was especially important where teams worked in different time zones or used different local versions of a handover form.
The third change was in sampling. A sampling rule may form part of a programme’s assurance model, but the sample cannot silently stand in for every unvisited door. The matrix states the coverage rule, the risk basis, who selected the sample, and what evidence remains required for the rest of the population. Planning Site Acceptance Criteria Across a Portfolio Refresh is useful for seeing that distinction at portfolio level.
The fourth change was handover continuity. At site one, the facility manager who attended commissioning could explain each decision. At site twenty, that knowledge was distributed. The acceptance record therefore carried enough context for an operator who had never visited the installation to understand why a decision had been made.
A multi-site handover checklist that can actually be audited
The multi-site handover checklist was arranged around the questions an auditor would ask later. Could the reader identify the door? Could they identify the rule version? Could they see what was observed? Could they see what remained open? The checklist did not ask the field team to recreate the whole programme; it asked them to leave an object-level trail that would survive a change of crew.
Before the site visit
Confirm the object schedule and use the door or unit identifier, not only a room label.
Record the model and variant from the label or catalogue row.
Confirm the matrix version, local annex, coverage rule, and exception route.
Assign the field tester, local confirmer, portfolio owner, and escalation contact.
Confirm the required record language and where the approved evidence will be stored.
At the door
- Record the date and time of the acceptance test.
- Describe the symptom or condition in the technician’s words and add the measured observation.
- Run the approved checks in order and record what each one ruled out.
- Record the credential state before and after where an access path was tested.
- Test an approved fallback route only where the specific SKU and matrix make it applicable.
- Record any part fitted or unit swapped, including the retained old unit where required.
Before release
- Attach the local evidence and name its authority and scope.
- Link every result to the correct door or unit.
- State the decision as accept, conditional accept, hold, or reopen.
- Give every open item an owner, required closure evidence, and controlled status.
- Obtain the facility manager sign-off only after the decision and evidence are visible in the same record.
After release
- Publish the approved record and the controlled matrix version.
- Carry unresolved exceptions into the operational handover.
- Link configuration changes, hardware swaps, and local annex changes to the affected doors.
- Preserve the original observation, supporting photo or reading, and technician notes.
The facility manager sign-off was therefore an accountable act, not an ornamental signature. It confirmed that the decision record was complete enough for operations. It did not certify a future outcome, settle an unresolved local requirement, or transfer responsibility for an unverified supplier claim.
Decision table: accept, condition, hold, or reopen
At scale, the programme needed a vocabulary that everyone could use without turning a discussion into a dispute over tone. The four states below separated evidence quality from physical condition. The decision table was short enough to use at the door and precise enough to audit later.
| Decision state | Evidence condition | Operational effect | Next owner |
|---|---|---|---|
| Accept | Common-core checks pass, required local evidence is attached, and no release condition remains open | Door record is released into the agreed operational handover | Portfolio acceptance register |
| Conditional accept | Core operation passes and only an expressly identified administrative or deferred item remains, with no unresolved door-operability or access-path failure | Door may operate only under the approved local arrangement while the exception remains visible | Named exception owner |
| Hold | A required check fails, critical evidence is absent, or the record cannot be linked to the correct object | Door stays outside normal acceptance and follows the approved local access procedure | Site lead and escalation owner |
| Reopen | A signed record is affected by a material configuration change, approved correction, or newly identified evidence conflict | Previous acceptance is suspended for review while unaffected records remain valid | Change controller and portfolio owner |
Conditional acceptance was the state most likely to be misused. It was not a soft accept. The matrix had to name the exact item, the interim operating condition, the owner, and the evidence required to close it. An open item could not migrate to another site’s list or become “complete” through verbal confirmation.
A hold did not automatically mean the lock was faulty. Missing evidence, an object mismatch, or an unresolved local annex could also produce a hold. Likewise, a physical pass did not erase an administrative failure. The decision followed the unmet condition rather than a general impression of the site.
This structure also supported site acceptance test evidence without making every evidence field compulsory in every situation. Mandatory fields were identified by the common core, the local annex, and the exception type. Optional fields were labelled optional for a reason. The Site Acceptance Criteria Documentation Set can hold the controlled forms, while the decision table explains how they operate together.
Day four: escalation without restarting every site
By day four, a failed check no longer triggered a full programme restart. The field lead classified the failed condition, preserved the observations already collected, and sent only the information needed for the next decision. The aim was to avoid both extremes: carrying a fault forward without evidence or discarding valid work because one item failed.
A door-operability failure went to the door-level fault procedure. The technician checked the physical and mechanical state before replacing a unit. An access-path failure was tested against configuration and credential scope before hardware action. A configuration mismatch was corrected and verified against the approved record. A missing document became an evidence exception rather than a claim that the physical work had failed.
The escalation packet carried the object identifier, model and variant, dated product-page reference, observed symptom, test sequence, what each check ruled out, affected credential state, photographs or readings, and the action already taken. It also carried the proposed decision. That let the reviewer answer a narrow question rather than reconstruct the visit from scratch.
Supplier escalation followed the same discipline. The buyer asked what capability or process the supplier could evidence for the specific configuration and did not turn the question into a promise. Ask for written scope, the dated product-page reference, any applicable process document, and the evidence required to close the exception.
CONFIRM-PER-SUPPLIER
No supplier statement was allowed to replace the door-level observation. Equally, the buyer did not ask the supplier to infer a site condition that the record had not captured. The supplier and the site team each answered the part they could evidence.
The border crossing inside the record
Cross-border service records created a second form of acceptance risk. The technical result might be sound, but its context could be weakened when the file moved between teams. The diary team treated the record as an evidence package with provenance rather than as a loose collection of files.
Every record carried the source site, local time zone, and time of observation. A second time reference could be added when a shared programme clock mattered. It named the language of the technician’s notes and the version of the translated field form. Where a translated term could change the meaning of a pass condition, the original controlled clause remained visible beside the translation.
The model and variant travelled with the result, together with the dated product-page reference used to confirm product-specific assumptions. The credential record carried the tested identifier and its state before and after the check, without turning the handover file into a place to store unnecessary sensitive information. The physical evidence carried a file reference that remained usable after the original device changed.
That last point still required a local answer.
CONFIRM-WITH-LOCAL-DATA-PRIVACY-LEAD
The portfolio team also checked whether evidence could move to the programme repository, how long it should remain available, and who could access it under the buyer’s own arrangements. No universal retention period or cross-border transfer rule was assumed.
A local annex explained any restriction. The field record could still be complete without exposing data beyond the approved operating need, but that balance had to be confirmed by the responsible local owner. Cross-border service records were strongest when the authority, purpose, and storage route were all visible.
Day five: ownership after the facility manager signs
Acceptance transferred a door into a governed operating population. It did not end the need for records, change control, or fault evidence. The field team therefore handed over more than a completion percentage or a list of installed units.
The central portfolio register carried the site, object identifier, model and variant, acceptance state, matrix version, local annex version, exception status, and release date. The local custodian held the physical or electronic evidence under the approved arrangement. The supplier escalation route remained linked to the unresolved technical items. Each role had a defined next action.
Configuration changes were linked to the affected door and the original acceptance record. Hardware swaps required a fresh observation and the required parts or retention record. A local annex revision required an impact decision against the affected objects. The portfolio did not reopen every door by reflex, nor did it assume an approved change had no effect.
Training linked to the handover. A local operator who had not attended commissioning needed to know how to use the operating procedure, how to raise a fault, and what record followed a reported event. The programme avoided presenting acceptance as proof that no further training would be required.
Service measurement also began from the same discipline. MTTR is the mean of resolution intervals across a defined period and population. Availability is operable time divided by scheduled time over the stated population and period. Neither measure, by itself, explains a tail, an unreported door, or the reason an event took time. The buyer sets the planning target and supplies the records; the supplier can be asked to support the measurement method.
That boundary was written into the handover pack. It prevented a clean acceptance day from being used as evidence about future service performance and prevented an early operational metric from being read as a promise.
Portfolio close-out: what operations should inherit
At close-out, the programme inherited a controlled handover system rather than a collection of site certificates. The master index linked every site to its matrix version, local annex, door-level record, decision, and unresolved exception. A portfolio review could see which doors were accepted, which remained conditional, and which still required a hold or reopened decision.
The closing pack included the common acceptance matrix, each local annex, the object schedule, the acceptance register, the exception log, the change record, signed facility manager decisions, technician evidence, supplier responses, and dated product-page references. It also included the field record template, so a later fault could follow the same evidence discipline rather than starting with a blank form.
The programme did not average away exceptions. A high count of accepted doors could not make an unresolved door disappear, and a local variation could not be generalised to every site. The register showed the exception where the decision remained visible.
That was the lasting change at scale. The lock was still tested at the door, and the evidence was still attached to a physical object. What changed was the control around the decision: one matrix, local annexes, named owners, versioned records, and a clear route from a failed check to closure. For the facility manager entering post-award operations, that is what made multi-site acceptance operable rather than merely documented.
The practical counterpart to this point is the smart lock mortise compatibility. That is the same question site acceptance criteria multi site has to answer here.A practical starting point is to take one existing handover sheet, divide it into the common core, local annex, exception schedule, and change record, then test the structure against two contrasting sites. Ask the programme team to review one completed door record against that structure and identify the first decision the current form cannot explain. The resulting matrix, evidence list, and escalation route can then become the controlled starting version for the portfolio.
Nothing is binding until a supplier prices it, so request a quote.

